From: Horatio Zhang Hongkun.Zhang@amd.com
Add ras_poison_irq and functions. And fix the amdgpu_irq_put call trace in jpeg_v4_0_hw_fini.
[ 50.497562] RIP: 0010:amdgpu_irq_put+0xa4/0xc0 [amdgpu] [ 50.497619] RSP: 0018:ffffaa2400fcfcb0 EFLAGS: 00010246 [ 50.497620] RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000000 [ 50.497621] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 [ 50.497621] RBP: ffffaa2400fcfcd0 R08: 0000000000000000 R09: 0000000000000000 [ 50.497622] R10: 0000000000000000 R11: 0000000000000000 R12: ffff99b2105242d8 [ 50.497622] R13: 0000000000000000 R14: ffff99b210500000 R15: ffff99b210500000 [ 50.497623] FS: 0000000000000000(0000) GS:ffff99b518480000(0000) knlGS:0000000000000000 [ 50.497623] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 50.497624] CR2: 00007f9d32aa91e8 CR3: 00000001ba210000 CR4: 0000000000750ee0 [ 50.497624] PKRU: 55555554 [ 50.497625] Call Trace: [ 50.497625] <TASK> [ 50.497627] jpeg_v4_0_hw_fini+0x43/0xc0 [amdgpu] [ 50.497693] jpeg_v4_0_suspend+0x13/0x30 [amdgpu] [ 50.497751] amdgpu_device_ip_suspend_phase2+0x240/0x470 [amdgpu] [ 50.497802] amdgpu_device_ip_suspend+0x41/0x80 [amdgpu] [ 50.497854] amdgpu_device_pre_asic_reset+0xd9/0x4a0 [amdgpu] [ 50.497905] amdgpu_device_gpu_recover.cold+0x548/0xcf1 [amdgpu] [ 50.498005] amdgpu_debugfs_reset_work+0x4c/0x80 [amdgpu] [ 50.498060] process_one_work+0x21f/0x400 [ 50.498063] worker_thread+0x200/0x3f0 [ 50.498064] ? process_one_work+0x400/0x400 [ 50.498065] kthread+0xee/0x120 [ 50.498067] ? kthread_complete_and_exit+0x20/0x20 [ 50.498068] ret_from_fork+0x22/0x30
Suggested-by: Hawking Zhang Hawking.Zhang@amd.com Signed-off-by: Horatio Zhang Hongkun.Zhang@amd.com Reviewed-by: Tao Zhou tao.zhou1@amd.com Signed-off-by: Alex Deucher alexander.deucher@amd.com (cherry picked from commit bc3e1d60f933f823599376f830eb99451afb995a) Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2612 Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c | 28 +++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c index 3129094baccc..230435d30deb 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c @@ -83,13 +83,13 @@ static int jpeg_v4_0_sw_init(void *handle)
/* JPEG DJPEG POISON EVENT */ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN, - VCN_4_0__SRCID_DJPEG0_POISON, &adev->jpeg.inst->irq); + VCN_4_0__SRCID_DJPEG0_POISON, &adev->jpeg.inst->ras_poison_irq); if (r) return r;
/* JPEG EJPEG POISON EVENT */ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN, - VCN_4_0__SRCID_EJPEG0_POISON, &adev->jpeg.inst->irq); + VCN_4_0__SRCID_EJPEG0_POISON, &adev->jpeg.inst->ras_poison_irq); if (r) return r;
@@ -186,7 +186,8 @@ static int jpeg_v4_0_hw_fini(void *handle) RREG32_SOC15(JPEG, 0, regUVD_JRBC_STATUS)) jpeg_v4_0_set_powergating_state(adev, AMD_PG_STATE_GATE);
- amdgpu_irq_put(adev, &adev->jpeg.inst->irq, 0); + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) + amdgpu_irq_put(adev, &adev->jpeg.inst->ras_poison_irq, 0);
return 0; } @@ -535,6 +536,14 @@ static int jpeg_v4_0_set_interrupt_state(struct amdgpu_device *adev, return 0; }
+static int jpeg_v4_0_set_ras_interrupt_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned int type, + enum amdgpu_interrupt_state state) +{ + return 0; +} + static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) @@ -545,10 +554,6 @@ static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev, case VCN_4_0__SRCID__JPEG_DECODE: amdgpu_fence_process(&adev->jpeg.inst->ring_dec); break; - case VCN_4_0__SRCID_DJPEG0_POISON: - case VCN_4_0__SRCID_EJPEG0_POISON: - amdgpu_jpeg_process_poison_irq(adev, source, entry); - break; default: DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n", entry->src_id, entry->src_data[0]); @@ -619,10 +624,18 @@ static const struct amdgpu_irq_src_funcs jpeg_v4_0_irq_funcs = { .process = jpeg_v4_0_process_interrupt, };
+static const struct amdgpu_irq_src_funcs jpeg_v4_0_ras_irq_funcs = { + .set = jpeg_v4_0_set_ras_interrupt_state, + .process = amdgpu_jpeg_process_poison_irq, +}; + static void jpeg_v4_0_set_irq_funcs(struct amdgpu_device *adev) { adev->jpeg.inst->irq.num_types = 1; adev->jpeg.inst->irq.funcs = &jpeg_v4_0_irq_funcs; + + adev->jpeg.inst->ras_poison_irq.num_types = 1; + adev->jpeg.inst->ras_poison_irq.funcs = &jpeg_v4_0_ras_irq_funcs; }
const struct amdgpu_ip_block_version jpeg_v4_0_ip_block = { @@ -677,6 +690,7 @@ const struct amdgpu_ras_block_hw_ops jpeg_v4_0_ras_hw_ops = { static struct amdgpu_jpeg_ras jpeg_v4_0_ras = { .ras_block = { .hw_ops = &jpeg_v4_0_ras_hw_ops, + .ras_late_init = amdgpu_jpeg_ras_late_init, }, };